 All right, folks. Super happy you all are here today. I'm excited to talk about this topic. And for those of you who have been to one of my talks in the past, this is going to be extremely different to how I usually talk. This is one of those race through a lot of slides really quickly kind of talks, or the Lessig style of approach, rather than talk about pictures for a long time. So thanks for coming, everybody. I am with Precona, and we do open source databases, database monitoring and management. Keeping open source open is kind of the mantra that we live by. We take all the open core models that write proprietary features and write an open source version of it, give it away, and then try and figure out how to make money somehow, despite all that. And by the way, since we're in the community track, I'm also hiring for a head of community. So if you're interested, feel free to let me know. All right, like I said, the race begins. So start the timer. All right, so I grew up in the 80s. I'm now over 40 years old, which to some of you may be a youngin', so others may be like who's that old fart? And back in the 80s, there were a lot of fun toys that you'd play with. Transformers was a really popular one. And so you'd put together your Christmas list. You'd say, ah, I want all the cool ones. Give me all these amazing Transformers. But, you know, not everybody's parents made a ton of money. So sometimes you might get something that wasn't quite right. You'd get the go-bot. Just to give you a sense of the difference between the Transformer and the go-bot. The Transformer looks pretty darn cool for the 80s. Imagine you're thinking that thing and you pop open the go-bot and you get this sort of a halfway version of it, not really getting the job done. Big disappointment. And that brings us back to open source, the topic of our talk today, right? When there's success, people try and take that model, use that model, abuse that model, redefine that model to mean whatever they want it to mean so that they can make a dollar. Success brings imitators. Here's another example. Look at Star Wars. Again, real popular back in the day, coming back around again. And then look at Turkish Star Wars. You can go Google this up. It's literally a stolen music, stolen video from the actual movie compiled into a new movie, right? Just ripping the model off and making something that's not as good. And sometimes even the starting point tries to take things a little bit too far, right? Could be jar jar, could be light bent, could go either way. So open source is one, but at the same time open source is under perhaps more threat than it ever has had, right? Because now it's successful. That means everybody's trying to jump on the bandwagon. And then you get investors coming into it, which puts a whole new level of stress and pain and pressure upon startups that are trying to figure out how to be the next big thing. Because now you're not only under your own pressure. You're under the pressure of somebody who's sitting on your board, who will fire your CEO if they're not tripling, doubling every single year. It's a lot of pressure to be under. Before I joined Precona, I was a docker, we felt that kind of pressure every single day. And they're seeing all the traction happening, right? They're seeing the downloads growing, the active users, the engagement statistics, doubling, growing multiple percents on a weekly basis. They see that traction happening and they say, wow, there's gonna be something amazing we could do with this. If we can just put in $30 million, it'll be amazing. And we can somehow monetize, right? Build the adoption and then monetize. That's the new way. But things aren't that easy. There is no build the adoption and then flip the switch and suddenly you're making millions of dollars. Or at least not one that works for everybody who cares. There is one that works for the investor. There's not always one that works for everybody else who's part of people who care. And finally, our users don't care about our investors. They don't care about our company. They care about themselves and their needs. They're grabbing Legos. I use the analogy all the time of, users are looking for like a Lego castle and they wanna grab it, have the instruction booklet, switch, rip, replace, to do whatever they wanna do with it. They're cobbling that together into a solution that meets their needs. And they also like variety. I used to work at a company called Redmonk and my co-founder there, James Governor, said, technology is a fashion industry. It is not about solving your problems. It is about trend-driven development. It is about conference-driven development. Any of you who've put together a talk on a deadline without knowing what that talk was gonna be about, you know what I'm talking about. And people show up because they're looking for what's new. We're now spending a lot of time talking about Kubernetes. Our own product even built upon Kubernetes. But the commercial market is a big question. The excitement is real. There's thousands of people, even more than 10,000 people showed up at I think the 2019 KubeCon. Might be that high again this fall, we'll see. But people are always moving on to the next big thing. Before Kubernetes, there was Mesos. Before Mesos, there was Docker. Before Docker, there was OpenStack. Before OpenStack, there was Eucalyptus. You can keep going back and back and back and people are jumping on to the new thing. There's always something different. Like I said, people get excited by the new things. They hop onto it. They try it. I mentioned conference-driven development, but there's also resume-driven development. People are trying to build out the new technology on their resume. They've got some shiny new object on there and makes them more attractive for new job prospects. People intrinsically are scratching their own itch. That's the open-source way, but it's also just the way people live their lives. They're building on their own incentives. They don't care about the company. They don't care about the investors. Then we have to figure out, now what? If that doesn't work, if we can't get people using our product for $1 billion, how about a trillion? How can we turn this from a product play into a platform play? How can you build a monopoly that people cannot resist that they have to have? Going back to that Lego metaphor, how can you not be selling the bricks but sell the base plate? Everybody's got to have a base plate. You can't build a cool Lego project without the base plate. Everybody now wants to have their technology, their startup become the new base plate. We're talking about all the different infrastructure, all the different platforms. Everybody uses a platform all the time. Our product has the word platform in it. It means something completely different than your product that has a word platform in it. So infrastructure has become increasingly that battleground. We're seeing these new trends come along, these new platforms come along on what feels like an almost weekly basis. And yet, how do we make a business out of that and how do we do so in a way that respects everybody involved? Well, we've been iterating and looking for somebody who's going to hate that term. There you go. We've been looking at ways to make a business that uses open source software for quite some time. And we started with the classic services model. You figure out, oh, I can make some open source software, give it away as a charity, build some adoption, and we'll sell some support, sell some services. That's what a lot of the earliest open source business is built around. That services models are very high churn. People just jump ship to another service provider. It's a commodity. And so it makes it really hard to build a successful growing business. If you're trying to build the next big thing, it's hard to do that when you're losing 20% of your customers every year. When your starting point is taking a few steps back from where you ended up and you have to have new sales replace all those lost customers every single time. So then people thought up this great idea. How about open core? How can we have an open source product or at least something that was called open source, but then we build on top of it to add something. And that's something typically turned into enterprise features for whatever that term means. Everything that any company running at a significant scale needs. You throw that into your proprietary add-on layer. It's like single sign-on. It's audit logging. It's regulatory compliance. It's all those sorts of things that CIO says they need. And that had some success. But it also generated a lot of frustration in the community. Because to do that as a company, a lot of times what some venture funded startup might do is say, oh well, we're going to put a lot of the features that you actually need to run it in production only in the proprietary set. Which is really annoying. So that means that the free open source project now has a lot less value. You can't do much with it anymore because it's only a toy. It's only suitable for experiments. At that point, what do you get? Another model as a service. And this is the one that's becoming increasingly popular over the past, I'll call it decade or so. The rise of cloud. You got databases as a service, infrastructure as a service, platform as a service, software as a service. And you can see that there's real value being delivered. And so this is one that works especially for companies that are targeting developers or technologists to deliver it as a service model and people can perceive that you're doing work on their behalf. And so they feel much more comfortable paying for it. One of the downsides though is that as a user you give up a lot of control. There's a lot of pieces behind the scenes that aren't open source. This is as an example, take a look at how Amazon might create a product. There's an open source project at the core that they've adopted and hosted. But what about all the management tooling behind it? What about all the automation, all the integrations? Those in many cases are not open source anymore. And so you can't take the open source project and host it yourself. There is no way to take that out and run your own options. So it's lost part of the value of an open source project being I can take that thing, I can stand it up myself and do stuff with it. It depends. Good question, right? Where does open core end? Does open core end at everything you need? That's where we started to get into some of the longer-term questions about something like SSPL as an example. Because then it does feel open core-ish in that sense. Of forcing you to open source everything needed to operate. Then there's another model. This is one that we're experimenting with and it's starting to emerge. It's kind of an interesting one I think if we look at the SRE trend, a lot of the conversations in that space are about toil. How do we make toil go away? How do we make that busy work of every day I have to do these things stop happening? So what if we could provide enhancements to people's productivity? Take away some of that toil, automate some of those daily jobs, automate checklists, that kind of thing, to make that go away. And this is in a way an extension of the as-a-service model to bring that to operating. You can do that on-premises too. Yes, but I don't want to pitch you right now. Percona Platform is our product. That's one of the things we do. And in short, how we do it is we deliver a lot of the automation as a service through APIs to wherever people are running their stuff. So if we take a look at are these models working? Is this successful? And we look at we look at the company names and we look at the financials. It's looking successful for the companies involved, right? Lots of IPOs, lots of funding, recent kind of down market aside. There's been a lot of good stuff going on. And yet, things aren't working quite as well as they appear to be working as you start looking deeper. If you take a look through some of the financials of a lot of these companies building upon open source, they don't look so great. Because every one of them is losing money in many cases, quite a bit of money. I'm not going to re-throw all the numbers here, but you can see even the most popular ones out there are losing massive amounts of money on a quarterly basis. If you summarize that you can pull out some of the public financials and start taking a look through them. You'll see that losing money is the way businesses tend to work today. And yet, that's become accepted by the market as normal and healthy. This works when you're venture-funded. Because you've got money to lose. It doesn't work if you're bootstrapped or customer-funded. Because the money's not there. The money only comes in when you create value for a customer. It doesn't show up when you've got somebody on the other side talking about traction and exponential growth curves and triple doubles. Yeah. It's always history doesn't repeat itself, but it rhymes. And a lot of these come back to companies focused on going public, generating money, returning investor-funded dollars, talking about shareholder value and talking about it getting there eventually. Eventually we'll flip the switch, we'll start to profit. What does that eventually mean? And it all started not in a bad way. Not in a bad way, but really with Amazon making this the popular model. Right? It's not about evil, but it's about companies that show long-term growth trends without making any money. Often it's because they are investing all of those potential profit dollars back into capital spend, back into new data centers, back into new software development, whatever else it might be. But when you look at the financials, they're aiming to avoid making money. They're aiming to optimize for growth. When does it end? When do you stop growing and say, oh, now it's okay to make profit? Why aren't you profiting all along? And while you're doing that, in many cases you're still optimizing for your investors, or you're optimizing for your owners, or you're optimizing for your executives. And you're doing so while not respecting your community and your users enough. Not thinking about your stakeholders and getting to profit is hard. If you look at all those success stories, none of them are profiting. Getting to profit is hard, and it's very hard when your incentives are almost set up the opposite. They're set up to optimize for traction and growth rather than for profit. If you take a look at companies using open source, most of them don't want to pay. It's a challenge with the freemium style of business model is that you have to accept the vast majority of your users are never going to give you a dollar. In fact, two thirds is even on the low end of what you might see in a lot of companies. You might expect for a freemium model that 90%, 95% of your users do not pay you a dollar. You got to figure out how do you monetize around that last five. So what does success look like? And for people who are building a community or building an open source project, that success looks the same for you as it does for your investors. If you think back to the first open source project you got involved in, why did you start? Why did you begin? Most people got started for scratching niche, solving their own needs, and they might have stuck around for the joy of it. I spent almost 15 years working on an open source project, and in the end, I didn't stick around because of the technology I stuck around for the people. It was the relationships I built along the way. But if you're at a company making money around open source software somehow, do you still have that love? Do you still have that joy? Do you still care about your community and the people you work with and collaborate with in the same way? Or are you so busy focused on the growth in the investors that you're not focusing on the users and building those relationships and a lot of this, again, is driven by the pressure coming from people who are providing the money to generate that non-profitable growth. So the question was monetizing at different levels. What do you feel meant by that? That is I'll call it the web company model is one of those approaches of like, you're Google, you're monetizing around search ads, you're not monetizing around Kubernetes at least until recently. But the old school web company model or LinkedIn and the open source projects they're working on, they're not monetizing anything to do with the technology it's a completely different product offering. And at the time, it was a lot of because of that freemium model and that freemium model gets even more tilted with open source software than it does with something like an Atlassian, for example. You might have, instead of 5 out of 100 users pay you, you might have 5 out of 1000. And that's hard to get around. And open source in many cases is I don't want to say becoming trivialized, but the definition and the value of it are being pushed to and beyond their limits. It's being used as a specific strategy, which has its merits. But it's then being taken and stretched to a point where the open source project itself has very little value. It has no standalone value in itself anymore. You say, oh, we've open sourced our SDK. What am I going to do with that? I can't do anything with your SDK besides use your proprietary cloud hosted software. It's not a valuable open source project in that way. It's a marketing device. Open sourcing your database drivers for different languages. They only work with your database. That's not the right use of open source. That's abusing open source. If the database is open source or if the core itself is not also open and also provides value running it in a production context or for real work. Otherwise, you're using it and abusing it. At the same time, we've got companies that need to develop something sticky. It can't just be, oh, use this open source commoditized thing, use this platform that works with everything because companies are trying to generate the same customers all the time. That drives them toward lock-in. How do you create lock-in? How do you get people to stick with you by making it hard to leave and doing so through technology rather than doing so through community and relationships? I'm not going to walk through this whole quote but here's an example of stickiness through database vendors because it's hard to migrate away. It's hard to switch. It is the stickier it gets but not for a good reason. The customer doesn't want to stay with you. They're forced to because you've raised the barrier so much to exiting. That doesn't generate a happy customer. That generates a pissed-off customer who's looking for an excuse to walk away and it leaves you to monetize through things like doing software audits. Over time, things keep changing of how do we try and make money around open-source software? We see these generational shifts of you try and force things to stay open, people shift the way they're delivering the product, you try and force them to stay open, people shift the way they deliver product. It kind of iterates over and over and over where the license is trying to chase after a way to force the code to stay open. Our latest point is the SSPL and similar variants which go after trying to force you to do nothing because you're so confused about what it's even asking of you. What can you do with Elastic Search Cloud Service today? If you open-sourced all of your management tooling, could you offer that? Can you answer that question? Can your lawyer answer that question? Not easy. I called it, as a service offering it's hard or impossible depending on your legal team and how you interpret stuff because if you read through some of the clauses in the SSPL they arguably force you to open-source all the way down to your infrastructure software and beyond. What about your hypervisor? What if you bought that stuff from a vendor? What if you're running on somebody else's cloud? Can you offer that as a service? Lots of questions. What if you're integrating with an SSO provider, like Active Directory? Can you open-source that? Right? You end up in a very sticky point. And yet at the same time the SSPL is starting to become outdated and it's doing so because everybody else is moving to the cloud now. And so as an investor or as a startup you're thinking, I'm going to offer my service as a cloud and I might not even have an open-source version of it at all anymore. It's just accessible through an API, so therefore I don't need open-source, I don't need the SSPL. But let's go back again to our users. Because this is, I think if you walk away from this thinking one thing, it's people besides the investors matter. And we have to keep that in mind. And we in this room as the community leadership conference have to represent the importance of community back in our own companies. And so when we talk about open-source for shareholders it's very different than open-source for stakeholders. What does our community want? They want to collaborate. They want freedom. They want to build upon the shoulders of giants so they can easily innovate for their own needs. Many of those things are exactly at odds with what a startup that's trying to generate revenue wants to happen with its product. They don't want you taking their product and freely innovating by modifying it, changing it, redistributing your own version of it. They want you to build upon their platform and generate lock-in. They want to have a playing field level so that they have a unique advantage over anybody else offering their version of the software. They want to have a CLA so that all of your ownership or non-exclusive license rights are signed to them so they can re-license whenever they feel like it. There's a lot of open-source projects out there without a company behind them. I did some analysis sometime back on GitHub projects. So I've got a question for you. What is the most common number of contributors in the world? One. And how often do you think that one person makes a living on that project? It's pretty darn close to zero. Unless you're the guy who works on Curl. There is an example out there. And so how do we help everybody generate a sustainable ecosystem around open-source? We just saw TideLift announced the series C today, maybe yesterday. So there are lots of experiments trying to figure out how do we create a healthy ecosystem that generates people's jobs, generates companies, but does so in a way that respects our users and helps them solve their problems. And again, going back to the earlier point, users don't really care about your company or your business or you need to generate revenue, they care about solving their problems. So they go through your buffet of all the free things, grab as much as they can eat, then grab a little bit more, pile it on, go back for seconds and thirds and fourths, and suddenly you're losing money and you can't figure out why. So it's open-source because you can't discriminate against field of use, because you can't limit the value that they're able to receive out of it and they don't feel any obligation to pay you. They're grabbing their Legos, they're doing stuff, but some of those Legos are a little defective. So you see vulnerabilities come out like log4j. And then you say, oh crap, now what? So we start up some critical initiative and say, oh, now we'll solve it. Well, for one example, two examples, three examples, getting that to scale is hard. In some cases, we see open-source maintainers that get so frustrated with the abuse as what they see of their trying to generate the community, trying to generate adoption, trying to generate a company around that, that they just kind of say screw you. They break their own library, they drop it from a repository and they go their own way. We see this a number of times in the MPM ecosystem but also in others. And so those kinds of breakages of critical infrastructure start to get the government involved because now that open-source is one, we have to figure out how to make it safe and how to make it secure for everybody. And so we see all kinds of executive orders coming in from the U.S. government as an example. We see all kinds of different government orders from other countries and around the world trying to say, here's what you have to do. And we see open-source package repositories enforcing new restrictions upon their maintainers. Saying you must now sign your code, for example. You cannot publish anymore unless you've signed your code and signed your release. Every one of those things creates new barriers and creates new concerns that if we don't self-regulate and do a better job ourselves, we're going to get the government stepping in and forcing us to do things which is going to be terrible for the community because it's going to chase more people away. And so what do we see happening in open-source across all these different areas? Why is there cause for concern? We're abusing open-source projects and open-source communities and our users in the sake of investor returns. And this is bad for us. This is bad for society. This is bad for our ecosystem. We take people who want to contribute and we say you cannot contribute because that competes with my proprietary feature of my open-core offering as an example. We're teaching new developers that open-source is just a money-making scheme. It's just a bait-and-switch trick but at some point we're going to take it away from you and say oh, yeah, the stuff you really need to run this is not open-source. Or we're going to change the license or we're going to change the pricing model like lightben, just the other day. And suddenly you thought you were safe but you're not. How do we defend ourselves by by insisting that we own our own code as one way? By not giving a single company unique rights and privileges to relicense as they see fit. We have things we can do within our own power to make a difference. We also need to think about new business models, new ways to structure the economics around open-source so that we respect all the stakeholders involved not only the shareholders. We need to think about how do we incorporate things like sweat equity into the ownership of open-source projects and open-source communities and companies that build upon them. How do we focus on open-source and success as something that is not only about revenue being generated but about community health as a whole and about value that's being created for the users of our software. One of the best ways to do that is to go back and apply for a leadership role at your own company. Get in there, get promoted, start running the place and you can push the incentives and you can push the metrics in the right way. The other way is if you've heard of Milton Friedman he's a well-known economist basically screwed all of us over about 50 years ago by saying shareholder value is the only thing that matters. Unfortunately some of that has started to come back around but it's people like us who need to push for stakeholder value not only shareholder value. The question comes back to each one of us of the choices we're making every day and what are we going to do about them? Are we going to be used and abused? Are we going to take advantage of our own user base? Or are we going to reach for the brighter side? Are we going to look for a better, happier place in the world idyllic as that sounds, as ridiculous as it might sound? We have to do what's right. We have to reinforce that we have to support others when we see them doing it and build that upward cycle instead of that downward spiral. So hopefully you enjoyed that. That is all I've got, thank you. I've got some time for questions if anybody has some. There is no difference, I've taken the other side of the spectrum where do you fall on that? So the question was where do I fall in that range around open source? Steve Wally, Jeff Borek have both taken different positions on this over the years. I see open source as a flavor an implementation of a freemium model in practice but it's one that also, I would say almost uniquely incorporates community in a way that most other freemium models cannot so if you imagine combining freemium with some kind of extensibility model and plug-in model that is effectively open source and yet community makes it I think uniquely different. So some more in the middle. So the question was more or less how do you align investor expectations with having open source software at all? Is it even possible? And I think the answer to that is it certainly is possible but it requires figuring out the right delivery model and doing so in a way that respects your users. The cloud hosting model is probably the most successful one right now because people see the value they're being created in it but what do you do when you start seeing some traction and then some other larger cloud provider comes along to post your own thing and make more money off of it? That's one of the biggest challenges right now for that approach. There are there's a whole school of thought around how do you build kind of differentiation that's sustainable and build some moats around the things that you're doing to keep that from happening effectively. I don't think licensing is the right way to do so. I think licensing is trying to apply some glue after you've already broken the table more or less. The right approach is more around building a community and making that sticky and using that community and that flywheel around things that the community creates and contributes to build something that sticks with you. I'll just to make that more specific as an example, I spent some time at Docker and one of the biggest reasons Docker is sticky is the content on Docker Hub the official images but also the ecosystem of things like partner images and community contributed images keeps it popular even when a lot of production environments have moved on to Kubernetes. Docker is still the developer tool of choice. I'll call on you in a second. Anybody else before Jeff asks another question? I have an online question. Oh, an online question, great. Thanks for this talk about generating revenue. What do you think about recognizing open source as a new type of digital public good and financing it through public money to address the free writer problem? For example add 1% open source tax to all proprietary software sales similar to value added tax and distribute the collected money back to the open source ecosystem based on specific usage and importance metrics. Super interesting question and a thought on how to approach that. I think that is the kind of model that has some potential here in Europe where people care about social safety nets and care about supporting the public good and is much less likely to be viable in the United States where it's more of everybody is on their own and best of luck to you kind of model. So there's some potential. Thank you. You said your company is doing something that you have not mentioned like how you are bringing in that model. Yeah, so I'll just give this as an example and try to avoid pitching anybody on it. If you feel like I'm pitching just go ahead and walk out the door. Totally fine with that. So we do something it's roughly like a red hat for databases is how I think about our business model. But part of that sort of hybrid RoboCop style thing that we do is one of these things we call advisors that are more or less an automated checklist and remediation tools for all kinds of common scenarios that we learn about based on supporting databases at scale. And so we deliver those through a called SAS but kind of like an infrastructure SAS sort of model where those get downloaded onto a production environment at our customers and users and they run on a regular basis and check how things are going recommend changes when things aren't going you know enable some sort of auto-remediation for customers and users who are willing to accept that kind of thing. And that's an example where if you stopped having those tomorrow your production environment would still keep working. But it would take you more work and you'd be more likely to miss stuff along the way. So it's a productivity enhancement and kind of a human augmentation approach Everything that runs on a production environment is open source. I hope they get frustrated if we're gone otherwise we don't have much of a business to build on. But you've got to create value somehow. And I think that how is increasingly if you want to have a business that grows it's got to be a software business somehow. So you have to figure out what is the most respectful way to your users in your community to create some kind of a viable company. I see the as a service model and some sort of an augmentation and productivity type of model as two of the ways that do that the best without taking things away or forcing things to be proprietary. Are there questions? Yeah in the back. So the question was you know what are some new business models that jump out especially for projects that may not be looking for investment and just have a couple of maintainers and are trying to figure out how do I build a healthy life around that. The tide lift one is one of the more interesting ones that I've seen lately you know we've seen a lot of people trying some sort of a sponsor patronage model over time and it hasn't really worked in a professional way for anybody. Nobody's really making a salary that they can live off of with that. And so the tide lift one is kind of interesting of trying to take subscription support and then spread it out across different open source projects. It does have some interesting incentives around if your project gets more bug requests do you get paid more money and therefore should you have like worse software. So you make more. But that one is interesting. I think another one you know even as a small company you can often figure out some way to offer a hosting model. So I definitely recommend thinking about how can you take your couple of people and make them a force multiplier. Right so it's not a couple of people doing a couple of support tickets. Is that at that kind of scale it doesn't work because you're always getting a phone call in the middle of the night with somebody who has this broken stuff. Percona started as an emergency consulting business back in the day 16 years ago now and moved on from that because it wasn't viable for the people involved. It also wasn't very predictable. So how do you figure out a predictable revenue model like some kind of a subscription that creates value for your customers and avoids locking them in because they don't want to be with you. It locks them in because you get extra value that you no longer get if you stop. So I think some sort of hosting combined with some sort of I'll call it slow support there's a tool that we use for community dashboards called Orbit and if you're an Orbit customer they give you email support after like two days. Which is extremely viable. It doesn't work for production software. Nobody's going to wait two days to be like I'm losing $900,000 a minute or something according to Gartner and I'm going to wait two days for this. That's not viable. That's some of my thoughts but I think I would love to hear more examples of smaller communities that have generated a great company around it. Downstream into that so obviously we all use those components down into that so you get that support or is that tomorrow a problem? So how do you feed back into the smaller communities and ones that I think might be a dependency of your product? It's a good question. It's the kind of thing where you would imagine some sort of a solution like micro payments to somehow work. I'm not aware of anywhere where it has worked in practice and so what is the approach and I think a part of it is raising the visibility of your project so that people even realize they're building upon you. For example if you're two levels away nobody knows you exist. You just get pulled in as part of some installation process with like 300 different libraries and toolkits and whatever else and then you're aligned in somebody's terminal output that they maybe read. How do you raise that visibility? It's a hard problem. You can do it by almost becoming infamous like Daniel the curl guy becoming your own XKCD comic and if you're core infrastructure every time you have a security vulnerability like it's terrible but it's also good because people now know you exist and know you matter and know they need to support you somehow but it's a tough problem. I don't have a good solution for you. Alright we're a couple minutes over. Last question. Why has Red Hat the exception that proves the rules to the question? Art. I can take a shot at that if you like. Go for it. He's got a mic coming your way one second. Sorry about that virtual attendee. One of the things was that operating systems were somewhat unique because of the need for certifications and so forth at the beginning and then I think just quite frankly and by the way I'm a Red Hat employee the but I think as time went on Red Hat just very much navigated this bringing value to customers and there were all kinds of timing things and we were over time so I will just shut off now. Yeah so lots of different reasons for it one that I think I heard from was that Brian Stevens some time ago was around simplifying the complexity of a massive ecosystem around the Linux distribution there aren't that many ecosystems that are complicated where you can create value off simplifying them. Thank you.